BUG:Sanity check on merge parameters for correct exception #26824

doc/source/whatsnew/v0.25.0.rst

@@ -130,7 +130,6 @@ than :attr:`options.display.max_seq_items` (default: 100 items). Horizontally,
 the output will truncate, if it's wider than :attr:`options.display.width`
 (default: 80 characters).
 
-
 .. _whatsnew_0250.enhancements.other:
 
 Other enhancements
@@ -156,6 +155,7 @@ Other enhancements
 - Error message for missing required imports now includes the original import error's text (:issue:`23868`)
 - :class:`DatetimeIndex` and :class:`TimedeltaIndex` now have a ``mean`` method (:issue:`24757`)
 - :meth:`DataFrame.describe` now formats integer percentiles without decimal point (:issue:`26660`)
+- :func:`pandas.merge` now raises ``ValueError`` when either of ``left_on`` or ``right_on`` is not provided (:issue:`26855`)
 - Added support for reading SPSS .sav files using :func:`read_spss` (:issue:`26537`)
 - Added new option ``plotting.backend`` to be able to select a plotting backend different than the existing ``matplotlib`` one. Use ``pandas.set_option('plotting.backend', '<backend-module>')`` where ``<backend-module`` is a library implementing the pandas plotting API (:issue:`14130`)
 - :class:`pandas.offsets.BusinessHour` supports multiple opening hours intervals (:issue:`15481`)
@@ -604,6 +604,7 @@ Other deprecations
  Instead, use :meth:`Series.dtype` and :meth:`DataFrame.dtypes` (:issue:`26705`).
 - :meth:`Timedelta.resolution` is deprecated and replaced with :meth:`Timedelta.resolution_string`. In a future version, :meth:`Timedelta.resolution` will be changed to behave like the standard library :attr:`timedelta.resolution` (:issue:`21344`)
 
+
 .. _whatsnew_0250.prior_deprecations:
 
 Removal of prior version deprecations/changes

pandas/core/reshape/merge.py

@@ -1091,13 +1091,19 @@ def _validate_specification(self):
  raise ValueError('len(left_on) must equal the number '
  'of levels in the index of "right"')
  self.right_on = [None] * n
+ if not self.right_on:
+ raise ValueError('both "right_on" and "left_on" '
+ 'should be passed')
  elif self.right_on is not None:
  n = len(self.right_on)
  if self.left_index:
  if len(self.right_on) != self.left.index.nlevels:
  raise ValueError('len(right_on) must equal the number '
  'of levels in the index of "left"')
  self.left_on = [None] * n
+ if not self.left_on:
+ raise ValueError('both "right_on" and "left_on" '
+ 'should be passed')
  if len(self.right_on) != len(self.left_on):
  raise ValueError("len(right_on) must equal len(left_on)")
 

pandas/tests/reshape/merge/test_merge.py

@@ -1026,6 +1026,22 @@ def test_validation(self):
  result = merge(left, right, on=['a', 'b'], validate='1:1')
  assert_frame_equal(result, expected_multi)
 
+ @pytest.mark.parametrize('merge_type', ['left_on', 'right_on'])
+ def test_missing_on_raises(self, merge_type):
+ # GH26824
+ left = DataFrame({
+ 'A': [1, 2, 3, 4, 5, 6],
+ 'B': ['P', 'Q', 'R', 'S', 'T', 'U']
+ })
+ right = DataFrame({
+ 'A': [1, 2, 4, 5, 7, 8],
+ 'C': ['L', 'M', 'N', 'O', 'P', 'Q']
+ })
+ msg = 'must equal'
+ kwargs = {merge_type: 'A'}
+ with pytest.raises(ValueError, match=msg):
+ pd.merge(left, right, how='left', **kwargs)
+
  def test_merge_two_empty_df_no_division_error(self):
  # GH17776, PR #17846
  a = pd.DataFrame({'a': [], 'b': [], 'c': []})
@@ -1763,3 +1779,20 @@ def test_merge_equal_cat_dtypes2():
 
  # Categorical is unordered, so don't check ordering.
  tm.assert_frame_equal(result, expected, check_categorical=False)
+
+
+@pytest.mark.parametrize('merge_type', ['left_on', 'right_on'])
+def test_missing_on_raises(merge_type):
+ # GH26824
+ df1 = DataFrame({
+ 'A': [1, 2, 3, 4, 5, 6],
+ 'B': ['P', 'Q', 'R', 'S', 'T', 'U']
+ })
+ df2 = DataFrame({
+ 'A': [1, 2, 4, 5, 7, 8],
+ 'C': ['L', 'M', 'N', 'O', 'P', 'Q']
+ })
+ msg = 'must equal'
+ kwargs = {merge_type: 'A'}
+ with pytest.raises(ValueError, match=msg):
+ pd.merge(df1, df2, how='left', **kwargs)